Fortifying angiogenic efficacy of conditioned media using phototoxic‐free blue light for wound healing

Abstract We used a blue organic light‐emitting diode (bOLED) to increase the paracrine factors secreted from human adipose‐derived stem cells (hADSCs) for producing conditioned medium (CM). Our results showed that while the bOLED irradiation promotes a mild‐dose reactive oxygen generation that enhances the angiogenic paracrine secretion of hADSCs, it does not induce phototoxicity. The bOLED enhances paracrine factors via a cell‐signaling mechanism involving hypoxia‐inducible factor 1 alpha. This study demonstrated that the CM resulting from bOLED treatment shows improved therapeutic effects on mouse wound‐healing models. This method contributes to overcoming the barriers to stem‐cell therapies, including the toxicity and low yields from other methods such as nanoparticles, synthetic polymers, and even cell‐derived vesicles.


| INTRODUCTION
Stem cell therapy has been widely used in wound healing through paracrine factors secreted from transplanted stem cells 1,2 While adult stem cells show low side effects (e.g., immune response due to their immunomodulatory properties) and do not raise ethical issues, 3,4 low viability of human adipose-derived stem cells (hADSCs) caused by harsh condition of damaged tissue remains as a problem that needs to be addressed. 5 Although materials ranging from scaffolds to nanoparticles have been used to increase the therapeutic effects of stem cell therapy, [6][7][8][9][10] abnormal immune responses attributed to low biodegradability and long-term side effects of these materials limit their bioapplication. 11,12 Moreover, the potential risk of carcinogenesis in the administering of stem cells is believed to be an obstacle to their uses in clinical settings. 13 To this end, conditioned medium (CM) promises to resolve the potential problem of transplant damage in stem cells. 14,15 The CM includes paracrine factors normally extracted from various cell cultures, 16 such as cytokines, growth factors, hormones, and other soluble factors that regulate cellular functions through ligand-receptor interactions via the cell signaling pathway. The application of CM ranges from osteogenesis, muscle regeneration, to ischemic disease. [17][18][19] In particular, the angiogenic efficacy of CM harvested from adult stem cells has drawn much attention due to its robust wound healing properties. [20][21][22][23][24] However, the low concentrations of therapeutic paracrine factors collected from normally cultured stem cells limit the widespread therapeutic usage. 14 While numerous attempts have been made to increase the therapeutic effects of CM by introducing external materials (including growth factors, hydrogels, and even designed biocompatible nanomaterials) to the stem cells, the high cost, process complexity, and nanomaterial contamination risks in CM still challenge its application. [25][26][27] Sung-Won Kim and Gwang-Bum Im contributed equally to this work.
Recently, the use of external stresses, such as light, acoustics, and shear stress instead of introducing external materials has gained much attention. 11,28,29 Among the external stimuli, light energy is widely accepted as a promising tool for wound healing. 30 In particular, red visible light (wavelength $ 650 nm) applied to wound areas in previous research 30 has been shown to enhance the angiogenic efficacy of stem cells, thereby showing stimulated therapeutic efficacy. 31 Furthermore, blue visible light ($420 nm) with high-density energy, has attracted attention as a promising tool for use in various fields, including soft robotics and cell sheet engineering. 32,33 However, to date, only a few studies have focused on blue visible light in therapeuticapplications because of its phototoxic effect on the cells, attributed to its reactive oxygen species (ROS)-generating property. 34 When cells are exposed, cytochrome c in mitochondria absorb the blue light, which disrupts electron transport and causes ROS generation. 35,36 The few existing studies include attempts to apply blue light to increase osteogenic 37 and angiogenic properties. 38 Blue light stem cell therapies, however, still need to be thoroughly explored in terms of cellular mechanisms and in vivo therapeutic efficacy before being used in clinical settings.
In this study, we used a blue organic light-emitting diode (bOLED) to increase the paracrine factors secreted from hADSCs, with controlled irradiation to reduce phototoxicity. Our findings elucidated the mechanism underlying redox signaling induced by mild ROS generation in which bOLED irradiation enhances the angiogenic paracrine secretion of hADSCs. We demonstrated the improvement of in vivo wound healing efficacy by injecting the CM extracted from bOLED light-treated hADSCs into the wound lesions of mice. Optimized condition of our bOLED-based CM therapy has a potential to replace the current clinical settings of cell therapy with advantages on cell function and CM manufacturing time and cost.

| Characteristics of bOLED and culture condition
In this study, we optimized the blue OLED light irradiation for hADSC culture to induce a mild-dose ROS generation (Figure 1a). We selected a bOLED with stable fixed values such as luminance of 43.78 Cd/m 2 , power efficiency of 8.03 lm/w, and peak wavelength of 472 nm wavelength when 4.5 V and 74 mA were applied. Figure 1b shows that the bOLED has a broad light spectrum biased towards green and red regions with a peak at 472 nm. To transfer the optimized energy to hADSCs, we irradiated the bOLED for 1 h 14 min, which had an optimized fluence of 22 J/cm 2 , as in our previous study. 39 (Figure 1c).
To determine whether our light system induces heat generation, we evaluated the expression of genes related to heat, such as heat shock protein (Hsp) 70 and Hsp 90α. While we found no significant difference between the two groups in Hsp 70 expression, a slight increase was observed in Hsp 90α expression in bOLED-treated hADSCs compared to hADSCs without bOLED treatment (Figure 1d).

| In vitro phototoxicity of hADSCs
High-energy blue light is known to induce phototoxicity in cells.
When cytochromes of the respiratory mitochondria absorb blue light, ROS is generated by disruption of electron transport ( Figure 2a). 32,33,40 High-energy blue light irradiation causes both endoplasmic reticulum (ER) stress and intracellular ROS levels in cells to rise, thereby increasing cell death-related genes to induce apoptosis. 41,42 However, we found that when OLED light with optimized light energy was used for irradiation, the actin filaments of hADSCs, which are cleaved during apoptosis, 43,44

| Angiogenesis improved by CM in an in vivo mouse wound model
To assess the therapeutic effects, we injected the above CMs into skin wounds on BALB/c nude mice models on Days 2-4 ( Figure 5a).   Similar to the representative images of wound-healing profiles, the wound closing ratio increased significantly in the BT group compared to the other groups (Figure 5c). The skin wound restoration was also observed in the CT group compared to the OT group, but the extent was lower than that in the BT group (Figure 5c). Figure 5d shows the representative vascular markers, including CD31 and Alpha-smooth muscle actin (α-SMA) exhibiting stronger fluorescent signals in the BT group compared to the other groups. Involucrin, a marker of epidermal differentiation in skin substitutes, was also increased in the BT group compared to both the OT and CT groups (Figure 5d). The total amount of CD31 and α-SMA gene expression levels in the wound region were significantly upregulated in the BT group compared to those in the OT and CT groups (Figure 5e, f). As shown in the H&E staining results (Figure 5g), the BT group showed enhanced reepithelization with increased thickness of the epidermis.

| DISCUSSION
Recently, tremendous efforts incorporating external materials into cells have been made to overcome the limited therapeutic effect of conventional cell therapy. However, although the transplanted cell survival rate and growth factor secretion were increased with incorporated materials, side effects, such as immune response limit widespread application. To this end, CM therapies have drawn attention as an alternative tool that can avoid immune side effects.
Numerous attempts have been made to increase the concentrations of growth factors by stimulating cell biomodulation. While light is a promising tool enhancing cell biomodulation, blue wavelength light has not been fully applied because of its ROSgenerating property, which can induce cell apoptosis. In this context, we have used low-energy-density, low heat-generating blue wavelength OLED light to generate mild-dose ROS in hADSCsinducing redox signaling to increase angiogenic growth factor secretion, while avoiding phototoxicity.
Since both mitochondria and NOX4 generate excessive amounts of ROS that can cause cellular damage when irradiated with highenergy blue light, we irradiated relatively low-energy blue light using OLEDs to induce redox signaling but without phototoxicity. Normally, OLEDs have low heat-generating properties, which are suitable for excluding heat-induced cellular damage during cell culture. We first confirmed that hADSCs treated with bOLED at 22 J/cm 2 do not show a substantial heating effect caused by the light source as compared to that of the NT group. A slight ROS increase was observed with DCFDA staining in the BT group, but apoptotic activity was not observed in the BT group compared to that in the NT group due to relatively low ROS generation.

| CONCLUSION
In this study, we present an effective method to improve wound healing with a CM treatment that can avoid the problems of conventional stem cell therapy. In addition, unlike conventional CM therapies, the CM obtained here using bOLED is capable of treating hADSCs with significantly higher amounts of angiogenic growth factors without phototoxicity. Although the detailed mechanism of blue light in stem cell therapy needs to be elucidated in future studies, we have introduced a method for increasing the angiogenic effects of CM using bOLEDs to target hADSCs.

| Cell culture
Human adipose-derived stem cells were purchased from Lonza (Basel, Switzerland) and cultured in DMEM (Gibco BRL, Gaithersburg, MD, USA) supplemented with 10% (v/v) fetal bovine serum (Gibco BRL) and 1% (v/v) penicillin/streptomycin (Gibco BRL). The cells were incubated at 37 C and 5% CO 2 saturation. The medium was changed every 2 days. Cells within seven passages were used in the experiments. For CM extraction, hADSCs were treated with Cu-AMN (Cu-AMN CM) or received no treatment (NT CM). For the protein assay, NT CM or Cu-AMN CM was extracted 1 day after treatment. The HUVECs were purchased from PromoCell (Heidelberg, Germany), and were cultured in endothelial cell media/media 2 (PromoCell) supplemented with Growth Medium 2 SupplementMix (PromoCell), and incubated at 37 C and 5% CO 2 saturation. The medium was changed every 2 days.

| Irradiation of bOLED light
Before irradiating hADSCs with blue light, the culture medium was replaced with a serum-free culture medium following phosphate bovine saline (PBS, Gibco BRL, NY, USA). Thereafter, the bOLED light was irradiated for 1 h 14 min, and a total of 22 J/cm 2 of light energy was transferred to the cultured hADSCs.  Olympus, Tokyo, Japan).

| Cell migration assay
The hADSCs were grown to confluence in 6-well plates. The hADSCs were then replenished with endothelial cell growth medium-2 (EGM-2, Lonza Bioscience, Basel, Switzerland) and irradiated with or without bOLED light for 1 h and 14 min at 22 J/cm 2 . A straight scratch was made on the layer of hADSCs using a P1000 pipette tip.
After incubating for 0 and 24 h, the gap width of the scratch repopulation was measured and compared to the initial gap size at 0 h.

| Western blot analysis
The hADSCs were collected and lysed in a radioimmunoprecipitation assay buffer (Rockland Immunochemicals Inc., Limerick, PA, USA).
After centrifugation at 10,000Âg for 10 min, the supernatant was prepared as a protein extract. The protein concentrations were deter-

| Human angiogenesis and cytokine array
Human angiogenesis (ARY007, R&D Systems, Minneapolis, MN, USA) was conducted using the manufacturer's protocol.

| Histology
The skin tissue specimens were retrieved 14 days post-treatment and fixed with a 4% formaldehyde solution. The samples were embedded in optimum cutting temperature (OCT) compound (SciGen Scientific, Gardenas, USA) and frozen to yield 10 μm sections. Hematoxylin and eosin (H&E) staining was performed to analyze the samples.

| Statistical analysis
All data are presented as mean ± SD. The statistical analysis was performed using GraphPad Prism (GraphPad Software, San Diego, CA, USA). To determine statistical significance, an unpaired Student's ttest was performed to compare two experimental groups, and ordinary one-way ANOVA was performed for the three experimental groups. Statistical significance was considered when the p-value was less than 0.05 or 0.01.